Mem0 Status Check

检查 mem0 本地记忆系统的功能完整性和运行状态。每次被问到 mem0 功能、状态、是否完整时触发。 触发词：mem0状态、mem0功能、检查mem0、mem0正常吗、mem0完整性

Security Scan

The skill's instructions require running local Python scripts and inspecting specific filesystem locations (hard-coded Windows paths) but the skill metadata does not declare those binaries or config paths — this mismatch is incoherent and could cause the agent to execute unexpected local code.

Purpose & Capability

The declared purpose is to check the local mem0 memory system, which reasonably requires access to local binaries/files. However, the skill metadata declares no required binaries, env vars, or config paths, while the runtime instructions reference absolute Windows paths (e.g., C:\Users\Administrator...\python.exe and D:\autoclaw\结果\mem0\mem0_wrapper.py) and a Chroma DB folder. The required local artifacts are not declared, which is inconsistent.

Instruction Scope

SKILL.md instructs the agent/operator to execute specific PowerShell commands that run local Python scripts, check a Chroma DB directory, and inspect cron/scheduled tasks. These instructions read and execute local files and rely on system state outside the skill package. They are not limited or parameterized and include platform-specific assumptions (PowerShell, 'cron' on a Windows environment), giving the agent broad discretion to run arbitrary local code.

Install Mechanism

There is no install spec (instruction-only), so nothing is written to disk by the skill itself. That reduces risk of the skill delivering code. However, because the instructions call local scripts and a specific Python executable, successful execution depends on pre-existing local artifacts; the skill does not declare or validate these dependencies.

Credentials

The skill declares no required environment variables or config paths, but the instructions implicitly require access to specific filesystem locations and a particular Python runtime. The omission of these requirements is disproportionate: the skill should declare the files, binaries, or credentials it needs, or parameterize paths rather than hard-coding them.

Persistence & Privilege

The skill is not always-enabled and is user-invocable. It does not request special platform-level persistence or claim to change other skills' configurations. No elevated platform privileges are requested in metadata.

What to consider before installing

This SKILL.md will attempt to run local PowerShell commands that execute a specific Python interpreter and mem0_wrapper.py at hard-coded Windows paths and inspect a D:\...\chroma_db folder. Before installing or enabling it: 1) Confirm you trust the files referenced (inspect D:\autoclaw\结果\mem0\mem0_wrapper.py and any called modules) because the skill's instructions will execute them. 2) Be aware the skill metadata doesn't declare these paths or required binaries — the author should instead parameterize paths or declare required config paths and Python. 3) Note the 'cron' check is ambiguous on Windows (verify whether the environment uses Task Scheduler or another scheduler). 4) If you do not control those files or cannot manually verify them, do not enable the skill. 5) To make the skill safer, request the author to remove absolute paths, add explicit dependency declarations (required Python runtime and config paths), and provide a way to run the checks manually or with explicit consent prompts.

mem0 状态检查

执行步骤

第一步：运行 mem0 search 测试语义搜索

& C:\Users\Administrator.DESKTOP-BD1JUD0\anaconda3\python.exe D:\autoclaw\结果\mem0\mem0_wrapper.py search main_user "test" --limit 1

预期：有返回结果（memory 字段非空）

第二步：运行 get_all 获取全部记忆数量

& C:\Users\Administrator.DESKTOP-BD1JUD0\anaconda3\python.exe D:\autoclaw\结果\mem0\mem0_wrapper.py get_all main_user

预期：返回记忆列表，计数 > 0

第三步：检查 cron 任务状态

# 从当前会话的 cron list 获取 mem0 相关任务运行状态

第四步：检查 Chroma 向量库

Get-ChildItem D:\autoclaw\结果\mem0\chroma_db -ErrorAction SilentlyContinue | Measure-Object

预期：文件数量 > 0

---

检查项目列表（逐项报告）

#	功能	状态	说明
1	add 第一次 LLM 提取	待确认	使用 USER_MEMORY_EXTRACTION_PROMPT + AGENT_MEMORY_EXTRACTION_PROMPT（含 few-shot examples）
2	add 第二次 LLM 对比	待确认	应有 ADD/UPDATE/DELETE 决策逻辑，目前疑似直接调 m.add()
3	search 语义搜索	待确认
4	get_all 获取全部记忆	待确认
5	update 更新记忆	待确认
6	delete 删除记忆	待确认
7	history 操作历史	待确认
8	delete_all 全量删除	待确认
9	reset 重置	待确认
10	流程记忆 Procedural	待确认	memory_type="procedural_memory" 支持
11	情景记忆 Episodic	待确认	cmd_add_episodic() 实现
12	MEMORY_ANSWER_PROMPT	待确认	cmd_chat() 里使用
13	增强版提示词模板	待确认	prompts_zh.py 中的模板
14	MCP Server (stdio)	待确认	mem0_mcp_server.py
15	定时去重合并 cron	待确认	mem0-daily-dedup，每天凌晨3点
16	Chroma 向量库存储	待确认	D:\autoclaw\结果\mem0\chroma_db

---

输出格式

【mem0 状态报告】

✅ 正常运行：
- search：正常
- get_all：X 条记忆
- 定时 cron：mem0-session-memory / mem0-daily-dedup
- Chroma：X 个文件

⚠️ 缺失功能：
- [具体功能名]：[原因/说明]

🔧 建议补充：
- [具体建议]